- Product Monk
- Posts
- How Space Could Make AI Computing 10x Cheaper
How Space Could Make AI Computing 10x Cheaper
Andre Borczuk
October 24, 2025 • Approx. Reading Time: 11 minutes
In partnership with
Partner Message
Find out why 100K+ engineers read The Code twice a week
Staying behind on tech trends can be a career killer.
But let’s face it, no one has hours to spare every week trying to stay updated.
That’s why over 100,000 engineers at companies like Google, Meta, and Apple read The Code twice a week.
Here’s why it works:
No fluff, just signal – Learn the most important tech news delivered in just two short emails.
Supercharge your skills – Get access to top research papers and resources that give you an edge in the industry.
See the future first – Discover what’s next before it hits the mainstream, so you can lead, not follow.
A new chapter in AI infrastructure is being written above our heads. Starcloud—once a Y Combinator startup, now a vanguard of orbital innovation—will next month launch the first commercial AI data center into low Earth orbit, powered by NVIDIA's H100 GPUs, marking the hardware's historic debut beyond the atmosphere. This feat signals not just a technical milestone, but the birth of a business model that could overhaul compute economics and global AI resilience.
Business Model & Scale
Starcloud's approach transforms traditional constraints. The heart of their proposition: solar-powered, GPU-rich data centers where electricity and cooling costs—the twin anchors of Earth-based facilities—are slashed by orders of magnitude. In practical terms:
Electricity costs are cut by an estimated 90% compared to terrestrial data centers, enabled by uninterrupted, atmosphere-free solar collection.
Cooling becomes almost trivial: the vacuum of space, at nearly -270°C, provides an "infinite" passive heat sink. Instead of massive chillers, Starcloud's architecture radiates heat away using one-meter black plates that deploy from the satellite chassis.
Each orbital payload is a modular "rack" of servers, storage, and networking, stacked on a rigid backbone, with a surface cloaked in solar panels capable of harnessing gigawatts of power over time.
AI Compute Power in Orbit
The significance of NVIDIA's H100 GPU debut in space cannot be overstated. These chips, regarded as the workhorse of deep learning and generative AI, power the largest language, vision, and simulation models on Earth. Starcloud's first 60 kg demonstrator, Starcloud-1, will lay the groundwork for orbital scale-out. Starcloud-2—scheduled for launch next year—will expand into edge AI and global cloud services, foreshadowing mesh networks of compute nodes circling the planet.
Starcloud's initial deployment will deliver 100x more GPU compute density per satellite than any previous space-based hardware, per company data.
Partner firm Crusoe is preparing to operate "the first public cloud in space," offering high-performance AI compute to commercial and, prospectively, defense customers.
Strategic & Economic Impact
Starcloud's orbital network is not a science experiment; it's a calculated answer to the exponential surge in AI workload demand that is straining terrestrial grids and physical infrastructure.
Key advantages:
Global reach and latency resilience: Low Earth orbit enables near-continuous coverage, creating a backbone for global AI services, disaster recovery, and mission-critical applications that must endure terrestrial outages or geopolitical shocks.
Scale and sustainability: With virtually unlimited solar input and passive cooling, these orbital centers avoid land and water use conflicts, sidestepping one of the costliest and most controversial issues facing today's hyperscale build-outs.
Economic argument: Starcloud claims, supported by industry analysis, that once launch amortization is included, total operating costs could run 10x lower than comparable high-performance data centers on Earth—a claim with major implications for hyperscale providers and energy planners alike.
Investment & Strategic Backing
Starcloud's journey began in 2024 as Lumen Orbit. Since then, the company has raised $24 million, drawing backing from Y Combinator, NVIDIA (through the Inception program), and national security-focused investor In-Q-Tel. The firm is led by CEO Philip Johnston, CTO Ezra Feilden, and chief engineer Adi Oltean—names poised to become synonymous with AI's new high frontier.
Technology Roadmap
November 2025: Starcloud-1 launches, targeting GPU-centric workloads in orbit.
2026: Starcloud-2 mission adds significant storage and networking capabilities, unlocking AI at the space "edge" and enabling robust, cloud-like services from every orbit.
Competitive & Policy Ramifications
With growing bottlenecks in power, water, land, and regulatory approval encircling terrestrial data centers, Starcloud's orbital approach offers policymakers and industrial players an escape hatch—and potentially a geostrategic advantage. The involvement of In-Q-Tel highlights early national security interest. If orbital data centers succeed, they could redefine data sovereignty, spur new international tech standards, and force a rethinking of spectrum allocation and orbital traffic management.
Market Outlook
Early industry modeling predicts that if Starcloud's cost and uptime claims withstand operational scrutiny, orbital AI compute could represent a multi-billion dollar addressable market before 2030, particularly as AI-reliant sectors (finance, logistics, defense, Web3) seek global-scale resilience.
Starcloud's orbital data centers are not mere technological feats—they're a harbinger of AI infrastructure unbound by the limits of power grids, land, or politics. As NVIDIA's H100s light up in the vacuum above, the rules of global compute may be about to change forever.
Are Your IT Costs Scaling Too Fast?
IT costs scale fast without standards. Deel’s free IT Policy Template helps reduce unnecessary spend, enforce best practices, and support compliance globally. Download it now to take control of your IT operations.
• Meta's new $1 billion El Paso data center marks a significant expansion in AI infrastructure, featuring a 1GW capacity specifically optimized for next-generation AI applications
• The facility is strategically designed to support Meta's ambitious AI-driven products, including smart glasses and real-time translation services
• The data center emphasizes sustainable practices while maintaining the computational power needed for large-scale AI operations
• This investment reflects the growing trend of tech giants building specialized infrastructure to support increasingly complex AI workloads
Why this matters for Product Leaders: Meta's aggressive data center expansion signals a pivotal shift in AI infrastructure requirements. As processing demands grow, product leaders must prepare for a future where cloud computing extends beyond Earth, potentially transforming how we architect and scale AI-powered products and services.
• AI industry leaders from OpenAI, Microsoft, and AMD advocated for increased AI infrastructure and chip exports during a US Senate hearing to maintain American technological advantage
• The debate centers on whether pursuing larger AI models is the optimal path forward versus exploring alternative approaches to achieve Artificial General Intelligence (AGI)
• This ongoing discussion highlights the critical strategic decisions facing the AI industry about resource allocation and research direction
• The hearing underscores growing concerns about maintaining technological leadership while balancing innovation with responsible development
Why this matters for Product Leaders:
The AI scaling debate raises fundamental strategic questions for product development. While infrastructure investments continue growing massively, the shift from pure computational power to meaningful intelligence suggests product leaders should focus on practical applications and user value rather than just raw model size.
• JPMorgan's ambitious $1.5 trillion initiative allocates $10 billion specifically to AI development, signaling a major vote of confidence in AI's role in financial services
• The investment strategy encompasses not just AI but also critical areas like cybersecurity and quantum computing, demonstrating a holistic approach to future-proofing operations
• This move positions JPMorgan at the forefront of financial technology innovation, particularly in using AI to enhance security and operational resilience
• The scale of investment reflects growing recognition that AI capabilities are becoming fundamental to maintaining competitive advantage in global finance
Why this matters for Product Leaders: JPMorgan's massive AI investment signals how financial institutions are positioning technology as core to security and business resilience. Product leaders should recognize this creates both competitive pressure and opportunity to integrate AI-driven security features into their roadmaps.
• OpenAI's collaboration with Walmart introduces seamless checkout functionality within ChatGPT, marking a significant advancement in conversational commerce
• The partnership enables users to make direct purchases through ChatGPT conversations, streamlining the traditional e-commerce journey
• This integration demonstrates how AI is transforming retail by merging natural language interaction with immediate purchase capabilities
• The development signals a broader trend of AI companies forming strategic alliances with major retailers to enhance customer experience and operational efficiency
Why this matters for Product Leaders:
OpenAI's Walmart partnership signals a pivotal shift in commerce-AI integration, where conversational AI becomes a direct sales channel. This creates urgency for product teams to rethink customer journeys and product discovery, as AI assistants increasingly influence purchasing decisions and shape consumer behavior.
UN-Limited Limit Orders
Why pay gas for limit orders that never execute? With CoW Swap, you can set an unlimited number of limit orders – more than your wallet balance – then cancel them all at no cost to you. Try Limit Orders.
Other Important News
Looking for more insightful reads?
Check out our recommendations that keep you updated on the latest trends and innovations across industries.
Wrapping Up
That's it for today's newsletter! How would you rate this edition?Please give detailed feedback so the next edition is even better! |
Looking for more insightful reads?
Check out our recommendations that keep you updated on the latest trends and innovations across industries.
Reply